There are several approaches of immunoprotection of pancreatic islets for the purpose of successful allo- or xenotransplantation in the absence of immunosuppressive medication. Extravasculair approaches are either macroencapsulation (large numbers of islets together in one device) or microencapsulation. The latter approach is to envelop each individual islet in a semipermeable immunoprotective capsule. Quite promising results have been achieved with polylysine-alginate microencapsulated islet grafts in rodents, but clinical application is still restricted to a very small number of cases. Relevant considerations regard the following aspects. The biocompatibility of the microcapsules is influenced by the chemical composition of the materials applied and by mechanical factors related to the production process. With purified instead of crude alginates, the percentage of capsules with fibrotic overgrowth is reduced to approximately ten percent, and the remaining overgrowth is mainly explained by mechanical factors, i.e. inadequate encapsulation of individual islets. Even with purified alginates, however, the duration of encapsulated graft function is limited to a period of six to twenty weeks. Obviously, other factors than bioincompatibility play a role, which factors have to be identified. The limited duration of graft survival cannot be explained by rejection since, in rats, survival times of encapsulated isografts are similar, if not identical, to those of encapsulated allografts. An important factor is probably insufficient nutrition as a consequence of insufficient blood supply of the encapsulated and thus isolated islet. This also influences the functional performance of encapsulated islet grafts. Although normoglycemia can be readily obtained in streptozotocin diabetic rat recipients, glucose tolerance remains severely impaired, as a consequence of an insufficient increase of insulin levels in response to intravenous or oral glucose challenge. Important factors are the characteristics of the capsules applied in view of optimal diffusion kinetics, and the fact that an encapsulated islet graft can only be implanted in the peritoneal cavity because of its volume. Further studies should focus on finding a practically applicable method to reduce the barrier between encapsulated islets and the bloodstream, in order to improve both the functional performance and the survival of encapsulated islet grafts.